Self-reported smoking cessation activities among Swiss primary care physicians

Physicians play an important role in smoking cessation, and medical advice and pharmacotherapy are effective clinical interventions to prompt and help patients to stop smoking [1‐4]. Brief physician advice is effective for successful smoking cessation with odds ratio (OR) 1.66 (95% CI 1.42 to 1.94) versus no advice or usual care [1]. The risk ratio (RR) of abstinence for any form of nicotine replacement therapy (NRT) relative to placebo or non-NRT group is 1.58 (95% CI 1.50 to 1.66) [3]. Bupropion double the odds of cessation (OR 1.94, 95% CI 1.72 to 2.19) [4]. Group therapy is also effective (OR 2.19, 95% CI 1.42 to 3.37 versus no intervention controls) whereas acupuncture, hypnosis, and autogenic training (relaxation technique with self-hypnosis) have no demonstrated efficacy on smoking cessation [2, 5‐7]. Self-help material such as brochures on tobacco and health might increase quit rates, but the size of the effect is small. Self-help materials have no additional benefit when used in concert with advice from a healthcare professional or with nicotine replacement therapy [8].

Around 80% of the Swiss general population visits a doctor annually, and most patients expect smoking cessation interventions from their physician [9‐11]. Furthermore, studies have found clinical activities related to smoking cessation to be highly cost-effective in developed countries, particularly when compared to other currently accepted preventive activities, such as treating mild to moderate hypertension and lowering cholesterol levels [12].

However, some doctors perceive barriers to engaging in smoking cessation counselling, such as lack of patient interest, lack of confidence in their ability to discuss smoking cessation, lack of time, fear of harming the doctor-patient relationship and difficulty to advice smokers without smoking-related problems [13‐16]. When practitioners are asked to judge the importance of different preventive activities in their clinical practice, 93% judge controlling blood pressure to be "very important"; in contrast, only 73% find counselling on smoking cessation to be "very important" [13].

Physicians' own health habits, like sedentary lifestyle or drinking more than three alcoholic drinks per day, could also negatively influence their attitudes toward smoking cessation counselling (smoking status was not a predictor in this study) [13].

Surveys of Swiss patients have reported that 81% to 88% of smokers have discussed their smoking habits with a physician, but only 34% were advised to stop [17, 18].

In 2002, when the survey was conducted, the national program for smoking prevention 2001–2005 has just started (Federal Office of Public Health). An educational program to improve physicians' skills and effectiveness in smoking cessation has been developed and assessed by residents in outpatients clinics few years before [19]. Twenty-five teachers were trained in 2002 in order to disseminate the smoking cessation program nationwide. A reference manual on smoking cessation in clinical practice including algorithms for smokers approach and therapy, as well as patients' brochures were available. There were however hardly any smoking cessation specialists, except in few University hospitals.

When the survey took place, smoking cessation post-graduate training sessions were not uniformly implemented throughout Switzerland with linguistic regions difference: there were proportionally more training sessions in French-speaking region than in German part.

The purpose of this survey was to assess clinical practice of Swiss primary care physicians toward patients motivated to stop smoking. Our ultimate objective was to highlight areas in which tobacco prevention programs should focus their attention in order to enhance delivery of evidence-based clinical practice for promoting smoking cessation.

The results of this survey suggest that the majority of Swiss primary care physicians practice evidence-based smoking cessation interventions. A large majority of these physicians reported prescribing first-line therapies including nicotine substitutes and bupropion, as well as providing individual counselling on how to quit using tobacco. Only a small minority reported recommending attending a smoking cessation group or referring smokers to tobacco specialists.

The only characteristic positively associated with a good practice score on smoking cessation intervention was having attended smoking cessation training. Physicians' own smoking status was not significantly correlated with smoking cessation practices, as has been reported in other studies; however, an association between personal health behaviour and attitude toward health promotion has been reported [13, 20, 21]. The interval since physicians received their medical diplomas and whether or not they practiced in an urban environment were not significantly correlated with tobacco cessation interventions.

International studies have assessed physicians' incorporation of tobacco control interventions in their clinical practice and their opinions about national guidelines [22‐25]. However, it was not possible to compare Swiss physician practices with this data because the questions in the surveys and the health systems are not similar.

Our study had some limitations. Firstly, our response rate (55%) led us to question whether there was potential bias. However, this participation rate is in concordance with rates reported in other physician surveys [21‐23]. Furthermore, response bias might not pose as much of a methodological challenge in this relatively homogenous population in terms of age and gender as compared to the general physician population, except the slight smaller proportion of respondents who were daily smokers, 7% versus 12% reported in a previous survey [26]. Secondly, respondents were perhaps more motivated in this topic, that could have resulted in overestimation of the quality of the smoking cessations interventions. A third limitation was using the score variables to combine responses to multiple survey questions. These unvalidated measures could have led to bias in the regression analyses. Giving equal weighting to the interventions is an approximate way to establish the score, as recommending counselling is probably a weaker intervention than direct counselling.

Just self-reported attitude, no measure of frequency, no measure of whether physicians assessed smoking status and non-uniformly implementation of smoking cessation training sessions with linguistic difference are additional limitations.

We think there are important policy implications in improving the practice scores among physicians who had or had not previously taken a smoking cessation course. Although this survey did not measure clinical outcomes in patients, we assumed that patients being followed by physicians with better smoking cessation practice scores would have higher quit rates. Therefore, our data provide impetus for promoting more smoking cessation training courses for physicians. Further evidence of the benefits of training physicians can be found in a prior randomized controlled trial among physicians, in which patients assigned to doctors who received smoking cessation training had improved smoking-related behavioural and clinical outcomes [27].

Since this survey was conducted in 2002, the Swiss national program for tobacco prevention published clinical recommendations and organized smoking cessation training courses for physicians [28‐31]. More than 4000 Swiss physicians have provided teaching assistance.

The majority of participating Swiss physicians practice recommended smoking cessation interventions with patients motivated to stop smoking. However, the number of physicians practicing these interventions could be increased and implementation of the interventions could be improved; smoking cessation training courses are an effective means of achieving both goals.

Department of Ambulatory Care and Community Medicine, Lausanne University Hospital, Switzerland. Institute of Social and Preventive Medicine (IUMSP), University of Lausanne, Switzerland.